Although it is generally acknowledged that longitudinal data provide the most information on growth and development and other time-dependent phenomena, such data are often analyzed by conventional (cross-sectional) statistical methods. This widespread practice ignores the distinctive characteristics (e.g., covariance structure) of longitudinal data and may yield misleading results. The purpose of this article is to present some strategies and make available computer programs for the appropriate analysis of longitudinal data. User-friendly PC programs for the estimation of average growth curves, computation of tracking indices, prediction of future values, diagnosis, classification, clustering, estimation of missing values, and testing hypotheses concerning individual and group differences are presented. Benefits of these methods over the usual techniques are illustrated with the example of maxillary growth in the rhesus monkey.
References
1.
Anderberg, M.R.: Cluster Analysis for Applications, Academic Press, New York (1973).
2.
Carter, R.L. and M.C.K. Yang: Large-Sample Inference in Random-Coefficient Regression Models. Comm. Stat. 8:2507-2526 (1986).
3.
Cohen, J.: A Coefficient of Agreement for Nominal Scales. Educ. Psych. Measur. 20:37-46 (1960).
4.
Cullis, B.R. and C.A. McGilchrist: A Model for the Analysis of Growth Data from Designed Experiments. Biometrics46:131-142 (1990).
5.
Dawson, D.V., A.B. Todorov, and R.C. Elston: Confidence Bands for the Growth of Head Circumference in Achondroplastic Children during the First Year of Life. Am. J. Med. Genet. 7:529-536 (1980).
6.
Edgington, E.S.: Randomization Tests, Marcel Dekker, New York (1987).
7.
Edlefson, L.E. and S.D. Jones: GAUSS (Computer Program), Applied Technical Systems, Kent, WA (1985).
8.
Fleiss, J.L.: The Design and Analysis of Clinical Experiments, Wiley, New York (1986).
9.
Foulkes, M.A. and C.E. Davis: An Index of Tracking for Longitudinal Data. Biometrics37:439-446 (1981).
10.
Friede, H., A.A. Figueroa, M.L. Neagele, H. Gould, N. Kay, and H. Aduss: Craniofacial Growth Data for Cleft Lip Patients Infancy to 6 Years of Age: Potential Applications. Am. J. Orthodont. Dentofacial Orthoped.90:388-409 (1986).
11.
Furey, A.M., C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: A PC Program to Aid in the Choice of the Design Matrix in Multiple Linear Regression. Int. J. Biomed. Comput.33:1-23 (1993a).
12.
Furey, A.M., T.R. Ten Have, C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: A PC Program Extending the Two-Stage Polynomial Growth Curve Model to Allow Missing Data. Int. J. Biomed. Comput.33:287-296 (1993b).
13.
Furey, A.M., C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: GTRACK: A PC Program for Computing Goldstein's Growth Constancy Index and an Alternative Measure of Tracking. Int. J. Biomed. Comput. in press (1994).
14.
Gauss (Computer Program): The GAUSS System Version 3.0. Aptech Systems, Inc., Maple Valley, WA (1992).
15.
Goldstein, H.: The Design and Analysis of Longitudinal Studies, Academic Press, New York (1979).
16.
Goldstein, H.: Measuring the Stability of Individual Growth Patterns. Ann. Hum. Biol. 8:549-557 (1981).
17.
Guo, I.Y., T.R. Ten Have, C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: A PC Program for Computing McMahan's Tracking Indices from One-Sample Longitudinal Data Sets. Int. J. Biomed. Comput.35:255-266 (1994a).
18.
Guo, I.Y., E.D. Schneiderman, C.J. Kowalski, and S.M. Willis: PC Program for Growth Prediction in the Two-Stage Polynomial Growth Curve Model. Int. J. Biomed. Comput.35:39-46 (1994b).
19.
Hills, M.: A Note on the Analysis of Growth Curves. Biometrics24:189-196 (1968).
20.
Hoel, P.G.: Methods for Comparing Growth Type Curves. Biometrics20:859-872 (1964).
21.
Kendall, M.G.: Multivariate Analysis, Griffin, London (1980).
22.
Kowalski, C.J.: Data Analysis in Craniofacial Biology with Special Emphasis on Longitudinal Studies. Cleft-Palate Craniofacial J. 30:111-120 (1993).
23.
Kowalski, C.J. and E.D. Schneiderman: Tracking: Concepts, Methods and Tools. Int. J. Anthrop.7:33-50 (1992).
24.
Kshirsagar, A.M.: Multivariate Analysis, Marcel Dekker, New York (1972).
25.
Luder, H.U.: Evidence for a Pubertal Spurt in Mandibular Condylar Growth of Nonhuman Primates. In: Craniofacial Growth During Adolescence. pp. 49-68. ( Carlson, D. and K. A. Ribbens, Eds.) Center for Human Growth and Development, Ann Arbor (1987).
26.
Mauger, E.A., C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: Extension of the Carter-Yang Growth Curve Model to Allow Unique Times of Measurement for Subjects. Int. J. Biomed. Comput. in press, ( 1994).
27.
McMahan, C.A.: An Index of Tracking. Biometrics37:447-455 (1981).
28.
Porter, C.L., P.H. Buschang, and E.D. Schneiderman: Studying Mandibular Growth with the K Tracking Statistic: Effects of Sample Size and Number of Tracks. J. Dent. Res.72(SI):340 (1993).
29.
Potthoff, R.F. and S.N. Roy: A Generalized Multivariate Analysis of Variance Model Useful Especially for Growth Curve Problems. Biometrika51:313-326 (1964).
30.
Rao, C.R.: Some Statistical Methods for the Comparison of Growth Curves. Biometrics14:1-17 (1958).
31.
Rao, C.R.: Some Problems Involving Linear Hypotheses in Multivariate Analysis. Biometrika46:49-58 (1959).
32.
Rao, C.R.: Some Observations on Multivariate Statistical Methods in Anthropological Research. Bull. Int. Stat. Inst. 38:99-109 (1961).
33.
Rao, C.R.: The Theory of Least Squares When the Parameters are Stochastic and its Application to the Analysis of Growth Curves. Biometrika52:447-458 (1965).
34.
Riolo, M.L., R.E. Moyers, J.M. McNamara, and W.S. Hunter: An Atlas of Craniofacial Growth, Monograph No. 2, Craniofacial Growth Series, Center for Human Growth and Development, University of Michigan, Ann Arbor (1974).
35.
Schneiderman, E.D.: A Longitudinal Cephalometric Analysis of Craniofacial Growth in the Rhesus Monkey (Macaca mulatta), Univ. Michigan Doctoral Dissertation, University Microfilms, Ann Arbor ( 1985).
36.
Schneiderman, E.D.: Facial Growth in the Rhesus Monkey: A Longitudinal Cephalometric Study. Princeton Univ. Press, Princeton, New Jersey (1992).
37.
Schneiderman, E.D. and C.J. Kowalski: Implementation of Rao's One-Sample Polynomial Growth Curve Model Using SAS. Am. J. Phys. Anthrop. 67:323-333 (1985).
38.
Schneiderman, E.D. and C.J. Kowalski: Implementation of Hills' Growth Curve Analysis for Unequal-Time Intervals using GAUSS. Am. J. Hum. Biol. 1:31-42 (1989).
39.
Schneiderman, E.D., C.J. Kowalski, and T.R. Ten Have: A GAUSS Program for Computing an Index of Tracking from Longitudinal Observations. Am. J. Hum. Biol.2:475-490 (1990).
40.
Schneiderman, E.D., S.M.Willis, T.R. Ten Have, and C.J. Kowalski: Rao's Polynomial Growth Curve Model for Unequal-Time Intervals: A Menu-Driven GAUSS Program. Int. J. Biomed. Comput. 29:235-244 (1991).
41.
Schneiderman, E.D., C.J.Kowalski, T.R.Ten Have, and S.M. Willis: Computation of Foulkes and Davis' Nonparametric Tracking Index Using GAUSS. Am. J. Hum. Biol. 4:417-420 (1992a).
42.
Schneiderman, E.D., S.M. Willis, C.J. Kowalski, and T.R. Ten Have: PC Program for Comparing Tracking Indices in Several Independent Groups. Am. J. Hum. Biol.4:399-401 (1992b).
43.
Schneiderman, E.D., S.M. Willis, C.J. Kowalski, and T.R. Ten Have: A PC Program for Growth Prediction in the Context of Rao's Polynomial Growth Curve Model. Comput. Biol. Med.22:181-188 (1992c).
44.
Schneiderman, E.D., C.J. Kowalski, T.R.Ten Have, and S.M. Willis: A PC Program for Computing the Foulkes-Davis Tracking Index for Polynomial Growth Curves. Int. J. Biomed. Comput. 32:35-43 (1993a).
45.
Schneiderman, E.D., S.M. Willis, C.J. Kowalski, and T.R. Ten Have: Longer-Term Growth Prediction Using GAUSS. Comput. Biol. Med.23:149-154 (1993b).
46.
Schneiderman, E.D., C.J. Kowalski, and S.M. Willis: Regression Imputation of Missing Values in Longitudinal Data Sets. Int. J. Biomed. Comput. 32:121-133 (1993c).
47.
Schneiderman, E.D., S.M. Willis, and C.J. Kowalski: PC Program for Estimating Polynomial Growth, Velocity, and Acceleration Curves When Subjects May Have Missing Data. Int. J. Biomed. Comput. 33:249-265 (1993d).
48.
Schneiderman, E.D., C.J. Kowalski, and S.M. Willis: Clustering on the Basis of Longitudinal Observations. Comput. Biol. Med. 23:399-406 (1993e).
49.
Schneiderman, E.D., S.M. Willis, C.J. Kowalski, and I.Y. Guo: A PC Program for Diagnosis on the Basis of Longitudinal Data. Int. J. Biomed. Comput.35:247-254 (1994a).
50.
Schneiderman, E.D., C.J. Kowalski, S.M. Willis, and I.Y. Guo: A PC Program for Computing Confidence Bands for Average and Individual Growth Curves. Comput. Biol. Med.24:119-127 (1994b).
51.
Schneiderman, E.D., S.M. Willis, C.J. Kowalski, and I.Y. Guo: Implementation of Exact and Approximate Randomization Tests for Polynomial Growth Curves. Int. J. Biomed. Comput.36:187-192 (1994c).
52.
Schneiderman, E.D., S.M. Willis, and C.J. Kowalski: A PC Program for Classification Into One of Several Groups on the Basis of Longitudinal Data. Comput. Biol. Med.24(4): in press (1994d).
53.
Stanek, E.J.: A Two-Step Method for Understanding and Fitting Growth Curve Models. Stat. Med. 9:841-851 (1990).
54.
Ten Have, T.R., C.J. Kowalski, and E.D. Schneiderman: PC Program for Analyzing One-Sample Longitudinal Data Sets Which Satisfy Rao's Two-Stage Polynomial Growth Curve Model. Am. J. Hum. Biol.3:269-279 (1991).
55.
Ten Have, T.R., C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: A PC Program for Performing Multigroup Longitudinal Comparisons Using the Potthoff-Roy Analysis and Orthogonal Polynomials. Int. J. Biomed. Comp.30:103-112 (1992a).
56.
Ten Have, T.R., C.J. Kowalski, E.D. Schneiderman, and S.M. Willis: Two SAS Programs for Performing Multigroup Longitudinal Analyses. Am. J. Phys. Anthrop.88:251-254 (1992b).
57.
Ten Have, T.R., C.J. Kowalski, and E.D. Schneiderman: A PC Program for Obtaining Orthogonal Polynomial Regression Coefficients for Use in Polynomial Regression Analysis. Am. J. Hum. Biol.4:403-416 (1993).
58.
Timm, N.H.: Multivariate Analysis with Applications in Education and Psychology, Brooks/Cole, Monterey, CA (1975).
59.
Vacek, P.M., R.M. Mickey, and D.Y. Bell: Application of a Two Stage Random Effects Model to Longitudinal Pulmonary Function Data from Sarcoidosis Patients. Stat. Med. 8:189-200 (1989).
60.
Vonesh, E.F. and R.L. Carter: Efficient Inference for Random-Coefficient Growth Curve Models with Unbalanced Data. Biometrics43:617-628 (1987).
61.
Ware, J.H. and M.C. Wu: Tracking: Prediction of Future Values from Serial Observations. Biometrics37:424-437 (1981).
62.
Zerbe, G. 0.: A New Nonparametric Technique for Constructing Percentiles and Normal Ranges for Growth Curves Determined from Longitudinal Data. Growth43:263-272 (1979a).
63.
Zerbe, G.O.: Randomization Analysis of the Completely Randomized Design Extended to Growth and Response Curves. J. Am. Stat. Assoc. 74:215-221 (1979b).
64.
Zerbe, G.O.: Randomization Analysis of Randomized Blocks Extended to Growth and Response Curves. Comm. Stat. A8:191-205 (1979c).
65.
Zerbe, G.O. and S.H. Walker: A Randomization Test for Comparison of Groups of Growth Curves with Different Polynomial Design Matrices. Biometrics33:653-657 (1977).
